GroupDocs.Parser is an on-premise document-parsing API that extracts text, formatted text (HTML, Markdown), tables, images, metadata, barcodes, hyperlinks, annotations and the table of contents from PDF, Word, Excel, PowerPoint, email, eBook, archive and image files - with OCR for scanned pages and template-based extraction of structured fields. SDKs for .NET, Java, Node.js and Python.

• NEW: GroupDocs.Parser for Python via .NET 26.10: engine 26.8 with the .NET 10 runtime embedded, Python 3.5 - 3.14 on Windows, Linux and macOS (Intel and Apple Silicon), annotation extraction, OCR for Arabic and Korean, and a groupdocs-parser command line. Download the wheels from releases.groupdocs.com; PyPI follows once it raises the package's file-size limit. Release notes.
• Published GroupDocs.Parser 26.10 for .NET on NuGet: a .NET Framework 4.8 build, external resources no longer loaded by default, and fixes for text extraction from some PDF and EML files. Release notes.
• The GroupDocs.Parser MCP server lets AI agents extract text, tables, images, metadata and barcodes locally, from a Docker image. Get started.

Extract text, tables, images and data from documents on .NET Framework 4.6.2+ and .NET 6, 8 and 10.

• GroupDocs.Parser-for-.NET: Core API examples for text, metadata, table, image and template-based extraction.
• Extract Annotations from PDF and PDF Table Extraction: use-case samples.

using System;
using GroupDocs.Parser;

// Extract the text of a PDF
using (Parser parser = new Parser("document.pdf"))
using (var reader = parser.GetText())
{
    // GetText returns null when the format does not support text extraction
    Console.WriteLine(reader == null ? "Text extraction isn't supported" : reader.ReadToEnd());
}

Native Java library for text, table, image and structured-data extraction.

• GroupDocs.Parser-for-Java: Java examples; the package com.groupdocs:groupdocs-parser comes from the GroupDocs Maven repository.
• Spring AI Document Reader and Document to Markdown RAG Chunks: use-case samples for AI pipelines.

import com.groupdocs.parser.Parser;
import com.groupdocs.parser.data.TextReader;

// Extract the text of a PDF
try (Parser parser = new Parser("document.pdf");
     TextReader reader = parser.getText()) {
    // getText returns null when the format does not support text extraction
    System.out.println(reader == null ? "Text extraction isn't supported" : reader.readToEnd());
}

Add document parsing to Node.js services and scripts with the Java engine: npm install @groupdocs/groupdocs.parser.

• Documentation | Release Notes

const groupdocs = require('@groupdocs/groupdocs.parser');

// Extract the text of a PDF
const parser = new groupdocs.Parser('document.pdf');
const reader = parser.getText();
// getText returns null when the format does not support text extraction
console.log(reader === null ? "Text extraction isn't supported" : reader.readToEnd());
if (reader !== null) reader.close();
parser.close();

A self-contained Python package with the .NET runtime embedded: pip install groupdocs-parser-net, Python 3.5 - 3.14.

• GroupDocs.Parser-for-Python-via-.NET: Runnable examples for every documented scenario, with a Dockerfile.
• PyPI | Docs | API Reference | Release Notes

from groupdocs.parser import Parser

# Extract the text of a PDF
with Parser("document.pdf") as parser:
    # get_text returns None when the format does not support text extraction
    text = parser.get_text()
    print(text if text is not None else "Text extraction isn't supported")

Let AI agents read documents through the GroupDocs engine, locally over MCP: no uploads, works in evaluation mode out of the box and with your existing license.

• GroupDocs.Parser.Mcp: The MCP server: extract text, tables, images, metadata and barcodes, and read document info.
• GroupDocs.Mcp.Installer: Cross-platform installer with per-client registration (Claude Desktop, Claude Code, VS Code, Cursor and more).
• GroupDocs.Parser.Mcp.Tests: Integration suite that verifies every published release.

# The server ships as a Docker image (linux/amd64 and linux/arm64); then ask your agent:
docker run --rm -i -v $(pwd)/documents:/data ghcr.io/groupdocs-parser/parser-net-mcp:latest

"Pull the line items table out of invoice.pdf as Markdown." "Are there any QR codes in shipping-label.png? If so, what do they decode to?"

MCP documentation | Docker | MCP Registry

• Invoice and receipt data extraction: pull totals, dates, vendors and line items with templates.
• Email and attachment parsing: extract headers, bodies, attachments and metadata from MSG, EML, PST and OST.
• Search indexing and RAG ingestion: turn mixed document collections into plain text, Markdown and tables.
• Contract review: read the comments and notes reviewers left, and the clauses around them.
• PDF table extraction: pull line items and financial tables out of PDFs.

• 65+ formats: PDF, Word (DOC, DOCX, RTF, ODT and templates), Excel (XLS, XLSX, XLSB, ODS, CSV, TSV and templates), PowerPoint (PPT, PPTX, ODP and shows), email (EML, EMLX, MSG, PST, OST), eBooks (EPUB, FB2, MOBI, AZW3, CHM), web (HTML, XHTML, MHTML), OneNote, Markdown, archives (ZIP, RAR, 7Z, TAR, GZ, BZ2) and images (PNG, JPG, BMP, GIF, TIFF, JPEG 2000).
• Plain and formatted text (HTML, Markdown), page by page or whole document.
• Tables, images, hyperlinks, barcodes, annotations, metadata and the table of contents.
• OCR for scanned pages and images, including Arabic and Korean.
• Template-based extraction of fields and tables; PDF form data.
• Containers: archive entries, email attachments and PDF portfolios.
• Text search with regular expressions, page-area extraction and page previews.

• .NET: dotnet add package GroupDocs.Parser
• Java: add com.groupdocs:groupdocs-parser from the GroupDocs repository, https://releases.groupdocs.com/java/repo/
• Node.js: npm install @groupdocs/groupdocs.parser
• Python: pip install groupdocs-parser-net
• Run the sample projects from the repositories above to extract text, tables, images and metadata.

• Documentation and tutorials: https://docs.groupdocs.com/parser/
• Free support forum: https://forum.groupdocs.com/c/parser
• Temporary license for full-feature evaluation: https://purchase.groupdocs.com/temporary-license

groupdocs-parser document-parser pdf-parser text-extraction data-extraction table-extraction metadata-parser email-parser ocr template-based-parsing content-ingestion rag python mcp mcp-server ai-agents